Content uploaded by Frederick de Serres
Author content
All content in this area was uploaded by Frederick de Serres
Content may be subject to copyright.
ORIGINAL ARTICLE
Abnormal overexpression of mastocytes in skin biopsies
of fibromyalgia patients
Ignacio Blanco &Nana Béritze &Mario Argüelles &Victoriano Cárcaba &
Fernando Fernández &Sabina Janciauskiene &Katerina Oikonomopoulou &
Frederick J. de Serres &Enrique Fernández-Bustillo &Morley D. Hollenberg
Received: 18 January 2010 / Revised: 17 March 2010 / Accepted: 20 April 2010
#Clinical Rheumatology 2010
Abstract Formalin-fixed, paraffin-embedded skin tissue
sections were collected from a matched cohort of 63
fibromyalgia syndrome (FMS) patients and 49 volunteers
from the general population with both alpha1-antitrypsin
(AAT) normal and deficiency variants. These tissues were
examined for the expression of the broad-spectrum inhibitor
AAT, the serine proteinases elastase and tryptase, the
proinflammatory cytokines MCP-1 and TNFα, the endo-
thelium biomarker VEGF, and the inflammation/nocicep-
tion-related receptor PAR
2
. The most relevant finding of the
study was a significantly increased number of mast cells
(MCs) in the papillary dermis of all FMS patients (greater
than or equal to five to 14 per microscopic high power
field) compared to zero to one in controls (p<0.001). MCs
strongly stained with tryptase, AAT and PAR
2
antibodies,
exhibited a spindle-like shape and were uniformly distrib-
uted around blood vessels and appendages. MCP-1 and
VEGF expressed weak/moderate positivity in most samples,
with a higher expression in controls than in FMS patients
(p<0.001 and 0.051, respectively). No differences in elastase
and TNFαwere found between both groups. Moreover, no
histological differences were found between samples from
I. Blanco (*):V. Cárcaba
Department of Internal Medicine, Valle del Nalón Hospital,
33920 Langreo, Principado de Asturias, Spain
e-mail: ignablanco@yahoo.com
V. Cárcaba
e-mail: victoriano.carcaba@sespa.princast.es
N. Béritze :M. Argüelles
Department of Pathology, Cabueñes Hospital,
Gijón, Principado de Asturias, Spain
N. Béritze
e-mail: nanarus@terra.es
M. Argüelles
e-mail: mario.arguelles@sespa.princast.es
F. Fernández
Department of Surgery, Valle del Nalón Hospital,
33920 Langreo, Principado de Asturias, Spain
e-mail: frfernando@telecable.es
S. Janciauskiene
Department of Respiratory Medicine, Hannover Medical School,
Carl-Neuberg-Straße 1,
30625 Hannover, Germany
e-mail: SabinaJanciauskiene@gmail.com
K. Oikonomopoulou
Department of Physiology & Pharmacology,
University of Calgary,
Calgary, Alberta, Canada
e-mail: oikoa@mail.med.upenn.edu
K. Oikonomopoulou
Department of Medicine, University of Calgary,
Calgary, Alberta, Canada
F. J. de Serres
Center for the Evaluation of Risks to Human Reproduction,
National Institute of Environmental Health Sciences,
MD-EC32, 79 T.W. Alexander Drive, Building 4401,
Research Triangle Park, NC 27709-2233, USA
e-mail: deserres@bellsouth.net
E. Fernández-Bustillo
Biostatistics Unit, Central University Hospital of Asturias,
33006 Oviedo, Principado de Asturias, Spain
e-mail: efbustillo@hca.es
M. D. Hollenberg
Department of Physiology & Pharmacology,
Medicine University of Calgary Faculty of Medicine Calgary,
Calgary, AB T2N 4N1, Canada
e-mail: mhollenb@ucalgary.ca
Clin Rheumatol
DOI 10.1007/s10067-010-1474-7
AAT deficiency and normal AAT phenotypes. Our results
indicate that FMS is a MC-associated condition. MCs are
present in skin and mucosal surfaces throughout the human
body, and are easily stimulated by a number of physical,
psychological, and chemical triggers to degranulate, releasing
several proinflammatory products which are able to generate
nervous peripheral stimuli causing CNS hypersensitivity,
local, and systemic symptoms. Our findings open new
avenues of research on FMS mechanisms and will benefit
the diagnosis of patients and the development of therapeutics.
Keywords Alpha 1-antitrypsin .Fibromyalgia .Mast cell .
Skin biopsy.Tryptase
Introduction
Fibromyalgia syndrome (FMS) is a prevalent idiopathic
disorder causing long-lasting, widespread soft-tissue pain
and generalized tender points identified by digit pressure on
the skin [1]. Other frequent associated complains are
abnormal fatigability, morning stiffness, non-restorative
sleep, chronic headaches, irritable bowel syndrome, urinary
frequency, temporo-mandibular joint disorder, psychological
distress and skin problems (i.e., dry, itchy, and tender skin),
etc. FMS is characterized by the central amplification of
sensory impulses (central sensitization) causing autonomic/
hormonal/immune/cytokine perturbations and interactions
[2]. Nonetheless, the source of stimuli and mechanisms
causing FMS central sensitization have not been clarified yet,
although emerging data suggest that it can result from the
disordered expression of inflammatory substances amplified
by genetic factors [3,4].
In the latest years, several morphological and immuno-
histochemical changes have been reported in skin biopsies
of FMS patients [5–12], including the presence of inflam-
matory cytokines [7], increased dermal IgG deposits [5,6],
overexpression and activation of extracellular matrix mast
cells (MCs), fibroblasts and mononuclear resident cells
[6,7], abnormal quantitative and morphological patterns of
dermal collagen [8,9], increased expression of nociceptive
glutamate N-methyl-D-aspartate subtype 2D (NMDA 2D)
receptors [10], and morphological changes of nociceptive C
fibers [11]. All these changes suggest that the skin is
injured in FMS patients (Table 1).
Human alpha1-antitrypsin (AAT) is a broad-spectrum anti-
inflammatory protein, highly abundant in blood and body
tissues. Its major function is the inhibition of over-expressed
neutrophil elastase and other serine proteinases [13,14]. AAT
is mainly secreted by the liver, and its gene has two major
groups of alleles, with each one named by the letters of the
alphabet. Normal alleles are called M. Thus, a normal
homozygous individual has an MM genotype. Most common
deficiency alleles are S (Glu2642Val) and Z (Glu342Lys). M,
S, and Z alleles express around 100, 40 and 15% of AAT,
respectively. Thus, mean serum levels for most common
AAT genotypes observed in clinical practice are: MM 100%
(95–200 mg/dL), MS 80%; SS 60%, MZ 55%; SZ 40%, and
ZZ 15%.
In recent years clinical, epidemiological, and pathological
evidence has suggested that inherited AAT deficiency might
play a role in the development of FMS, probably because of
the loss of AAT anti-inflammatory efficacy [3,15–17]. Thus,
the analysis of inflammation-related markers in FMS cases
with congenital AAT deficiency can provide a valuable model
to study how inflammatory substances may complement
genetic factors leading to the development of fibromyalgia
[16]. Seeking these markers, we analyzed the expression of
the broad-spectrum serine proteinase inhibitor ATT, the
serine proteinases elastase and tryptase, the proinflammatory
cytokines monocyte chemoattractant protein (MCP-1), and
tumor necrosis factor alpha (TNFα), the endothelium
biomarker vascular endothelial growth factor (VEGF), and
the pain/inflammation-related proteinase-activated receptor 2
(PAR
2
) in formalin-fixed, paraffin-embedded skin tissue
sections of a matched cohort of FMS patients and controls
collected from the general population (GP) with both AAT
normal and deficiency variants.
Methodology and results
Subjects
Sixty-three FMS patients (62 females and one male)
fulfilling the 1990 American College of Rheumatology
(ACR) established criteria [1] and 49 non-family-related
age-matched healthy volunteers (46 females and three
males) were enrolled for the present study. Most of these
112 subjects belonged to the two aforementioned large
cohorts of FMS patients and GP from the central region of
Asturias, Northern Spain [4].
The study project was approved by the Ethical Committee
of the Central University Hospital of Asturias, and endorsed
by the Spanish National Health Institute Carlos III (Ministry
of Health and Consumption, Madrid, Spain, project
PI061798) and the Biomedical Research Office (OIB) of
Asturias. All subjects signed an informed consent to
participate in the present study. Every FMS subject completed
a Spanish version of the fibromyalgia impact questionnaire
(FIQ), with a range of 0 to 80 points, and the Health
Assessment Questionnaire, with a range of 0 to 3 points.
The tender points (TePs) of FMS subjects were identified by
digit pressure on the 18 TePs locations recommended by the
1990 ACR study [1]. AAT serum concentrations and AAT
phenotypes were carried out in the AAT reference laboratory
Clin Rheumatol
of the National Silicosis Institute in Oviedo, Asturias (Spain).
AAT serum concentrations were measured by nephelometry,
and phenotyping was carried out by isoelectric focusing.
Biopsies The majority of the open biopsies were performed
by the same surgeon (FF) in an aseptic operating room at
the Valle del Nalón Hospital. Three of our fibromyalgia
female patients, who were also AAT-deficient, and one
AAT-deficient male subject, who was not affected by FMS
and was part of our control group, were treated with AAT-
replacement therapy. All four subjects accepted to delay
scheduled AAT infusions 2 weeks prior to their scheduled
biopsies. The majority of the remaining FMS patients
followed their daily pharmacological therapy as prescribed.
All samples were obtained from the external upper quadrant
of the left gluteal region, out of the ACR-1990 established
tender points [1]. About 4–6 ml of intradermal lidocaine
2% was injected 5–10 min before the incisions. Sample size
obtained oscillated between 8–18 mm in length, 4–7mmin
width, and 4–8 mm in depth. All skin samples obtained
included the epidermis, dermis, and part of the subcutaneous
adipose tissue, and were longitudinally hemi-sectioned. One
part was immediately frozen at −80°C and the other routinely
processed and placed inside a paraffin block. Paraffin blocks
were tag-labeled and sent to the Pathology Service of
Cabueñes Hospital (Gijón, Asturias) for a blind analysis by
two external pathologists (NB and MA).
Immunohistochemistry
A paraffin block of skin from each case was selected for
immunohistochemistry (IHC) at Cabueñes Hospital. Sections
of 3-μm thickness were cut from paraffin blocks, floated
onto salinized slides and dried for 30 min at 70°C and
overnight at 60°C, followed by deparaffinization in xylene
and rehydratation through graded ethanol. Skin sections were
stained with the following commercial antibodies: alpha1-
antitrypsin (anti-human alpha-1-antitrypsin, polyclonal rabbit;
Dako); elastase (anti-human neutrophil elastase, monoclonal
mouse, clone NP57; Dako); tryptase (anti-human mast cell
tryptase, monoclonal mouse, clone AA1; Dako); monocyte
chemoattractant protein (anti-human monocyte chemotactic
protein-1, MCP-1, monoclonal antibody, clone 24822, isotype
mouse IgG1; Leinco Technologies, Inc.); tumor necrosis factor
alpha (goat anti-human Tumor Necrosis Factor-a, TNFα,
polyclonal antibody; Leinco Technologies, Inc.); vascular
endothelial grow factor (anti-human Vascular Endothelial
Growth Factor, VEGF, monoclonal mouse, clone VG1; Dako);
and proteinase-activated receptor 2 (rabbit polyclonal to PAR
2
,
isotype IgG; Abcam plc). Sections were counterstained using
Table 1 Pathological changes described in skin biopsies of fibromyalgia syndrome patients (FMS) patients
Author year
[reference]
Study details Findings Author’s comments
Enestrom et
al. 1997 [6]
Case-control study: FMS #: 25,
controls #: 22. Punch biopsies
in flexor surface of forearm.
IMF and EM study
Significantly higher IgG deposits in dermis
and small vessels, higher reactivity to
collagen III, and increased number of
activated mast cells
in FMS
Neurogenic inflammation in
skin of FMS patients
Salemi et al.
2003 [7]
Case-control study: FMS #: 53,
controls #: 10. Punch biopsies in
deltoid. PCR and IHC
Positive signals of IL-1β, IL-6 and TNFα
in fibroblast and mononuclear cells at the
site nerves in up 38% of FMS samples
Inflammatory cytokines in FMS
suggesting an inflammatory
component in pain induction
Sprott et al.
2004 [8]
Case-control: 8 FMS vs. 8 healthy
controls. Trapezium (TePs). EM
Subepidermal concentric layers (“cuffs”)of
collagen around peripheral sensitive axons in
all FMS samples, not observed in any control
Neurogenic inflammation and ECM
collagen remodeling suggested
Ribel-Madsen
et al. 2005 [9]
Case-control: FMS #: 27,
controls #: 8. Samples from
thigh region. High performance
liquid chromatography. Light
and EM
Lamellar structure of perineurium and collagen
packing deficiency in endoneurium more
frequent and to a larger extent in FMS
In FMS the amount of collagen may
be lower and collagen packing less
dense in endoneurium
Kim et al.
2005 [10]
Case-control: FMS #: 11,
controls #: 8. Punch biopsies
in deltoid region. IHC
Increased expression of nociceptive glutamate
N-Methyl-D-aspartate subtype 2D (NMDA
2D) receptors in FMS
NMDA2DR, critically involved in
neuronal hyperexcitability, may
contribute to peripheral sensitization
in FMS
Kim et al.
2007 [12]
Case-control: FMS #: 13,
controls #: 5. Punch biopsies
in deltoid region. EM
Abnormal morphological patterns of
unmyelinated C fibers and associated
Schwann cells, with most cells being
ballooned and axons peripheralized in
Schwann cell sheaths
These anatomical abnormalities
may contribute or be due to
lower pain threshold in FMS
IMF immunofluorescence, EM electron microscopy, PCR polymerase chain reaction, IHC immunohistochemical staining, ECM extracellular
matrix, TePs tender points
Clin Rheumatol
hematoxylin. As a negative control, the primary antibody was
omitted and replaced by the Universal Negative Control
(UNC) Cocktail of mouse IgM and the four subclasses of IgG
(IgG1, IgG2a, IgG2b, and IgG3; Code IR750, DAKO) a
cocktail of immunoglobulins commonly used as a negative
control when performing in vitro diagnostic procedures.
Pathologists were given coded samples from each cohort
for analysis. Thus, every slide was first analyzed independently
by two pathologists (NB and MA), and the final report of each
case was issued after discussion and agreement between both
of them. Sequential magnifications of 2.5, 10, 25, and 40 (×10)
were used. Low magnifications were employed to select fields
showing about five dermal capillary vessels.
A hematoxylin/eosin (HE) magnification of 400, averaging
five fields per sample, was used for cell counting. Very weak
staining, which was faint or barely discernible,is denoted as ±;
weak staining, which was difficult but objectively recogniz-
able, is reported as +; moderate staining, which was easily
recognized, is shown as ++; and strong staining, with an
intense dark brown color, is indicated as +++. Cell staining
was further categorized as negative or positive, and the latest
was subsequently divided into focal (<50% cells being
positive) or diffuse (>50% cells being positive). To estimate
the inter-rater reliability between two raters, the Cohen's
kappa (κ) coefficient was used. When both raters were in
complete agreement we considered κ=1, while when there
was no agreement κ≤0. Using this range of values (0–1), the
following scale was applied: 0 no agreement; 0.1–0.20 slight
agreement; 0.21–0.40 fair agreement; 0.41–0.60 moderate
agreement; 0.61–0.80 substantial agreement; and 0.81–1.00
almost perfect agreement.
Statistics
The statistical analysis of demographic and clinical character-
istics of the cohort was performed by using SPSS 12.0
Statistics package. Descriptive statistics were used for primary
cohort database. Quantitative variables were expressed as
means with standard deviations and range. Groups were
compared by using the Mann–Whitney and Kruskal–Wallis
non-parametric test. Qualitative variants were compared by
means of contingency tables. A multivariate general linear
model was employed to compare the influence of two factors
(for example, AAT phenotypes in FMS and controls). The
value of p<0.05 was accepted for statistical significance.
Results
Demographic and clinical characteristics
One hundred-eight (96.4%) out of the 112 subjects studied
were females. The mean age of FMS and GP subjects was
53.4 (8.7) and 51.1 (7.5) years, respectively (Table 2).
Sixty-six out of 112 AAT genotypes (58.9%) were normal
MM variants, while 46 (41.7%) were MS, SS, MZ, SZ and
ZZ AAT deficiency variants (Table 3). Statistical analyses
did not show significant differences in ages, genotypic
composition, or AAT serum concentrations of patients and
controls. FMS subjects reported chronic pain with a mean
duration of around 10 years and showed high scores in
TePs, FIQ, and HAQ questionnaires [i.e., 16..4 (1.2), 63.8
(2.6), and 1.7 (0.2), respectively], being in general
categorized as moderate-to-severe FMS patients (Table 4).
In the FMS cohort, the prevalence of anxiety, insomnia, and
depression requiring pharmacological treatment was 60%,
40%, and 30 %, while one of these three conditions was
reported in about 5% in GP controls. As expected in
cohorts with a high number of AAT deficiency subjects,
airways chronic obstruction requiring daily pharmacologi-
cal treatment was detected in around 7% FMS and 5% GP
subjects. Less frequent co-morbidities (up to about 4%),
which were found in both groups, included dyslipidemia,
arterial hypertension, hypothyroidism, and diabetes, without
statistical differences between patients and controls. Most
FMS patients used some combination of non-pharmacological
(physical and cognitive) and pharmacological therapies,
including benzodiazepines (43 cases), non-steroidal anti-
inflammatory drugs (28 cases), omeprazole (30 cases), para-
cetamol (23 cases), serotonin uptake inhibitors (18 cases),
tryciclic antidepressants (14 cases), opiates (16 cases),
antiepileptics (six cases), metamizole (four cases), ACE
inhibitors (three cases), and other drugs such as bronchodila-
tors, vitamins, calcium, and diuretics (12 patients). As
mentioned above, three FMS/AAT-deficient patients and one
COPD non-FMS/AAT-deficient control received long-term
AAT-replacement therapy.
Skin biopsy samples, gross, and microscopic findings
Macroscopically, no significant differences were visible in all
samples, and both the gross morphology and the conventional
hematoxylin/eosin stain of all samples were accepted as
normal.
Immunohistochemical findings
Alpha1-antitrypsin (AAT) staining diffusely marked masto-
cytes of papillary dermis with moderate/strong intensity in
both FMS and control samples and also occasionally
stained endothelium cells and appendages focally. In 98%
of skin samples from controls, zero to one mastocytes
positive for AAT (AAT
+
) per microscopic high power field
(×400) were identified, while five to eight AAT
+
masto-
cytes per high power field were found in the remaining 2%.
In contrast, in 100% of FMS samples, five to 14 AAT
+
Clin Rheumatol
mastocytes per high power field were found, 47.6% of them
showing a perivascular location around small vessels, and
52.4% located around appendages or other papillary dermis
components (p<0.001).
Tryptase diffusely stained mastocytes of papillary dermis
with moderate/strong intensity, both in FMS and in control
samples. In 100% of controls, only zero to one mastocytes
positive for tryptase (tryptase
+
) per field were identified. On
the contrary, in 100% of FMS samples, five to 14 tryptase
+
mastocytes per microscopic high power field were found,
showing a perivascular location in 39.7%, and around
appendages or other papillary dermis components in 60.3%
(p<0.001).
A reverse pattern characterized the control staining of
PAR
2
.PAR
2
-negative mastocytes were seen in 57% of the
cases, while positive mastocytes (PAR
2+
)weredetectedin
43% of cases. PAR
2+
mastocytes were located around small
vessels in 31% of cases and in appendages or other papillary
dermis components in the remaining 12%. In FMS patients,
the percentage of patients with PAR
2+
mastocytes increased
to 83% of cases while 18%, of patients were negative. Of the
positive cells, 65% of PAR
2+
mastocytes were distributed
around small vessels and 18% were found in appendages and
other papillary dermis components (p<0.001). In addition,
focal positivity for PAR
2
was occasionally found at sites
distinct from mastocytes in epidermis, endothelium, appen-
dages, small muscle, and nerve tissues of the dermis, both in
controls as well as in FMS patients.
Elastase staining revealed a similar pattern between patients
and controls, as staining wasnegative in 84% of controls and in
89% of FMS patients. Focal weak or moderate elastase-
positive areas were found in epidermis, basal membrane, and
extracellular dermis of 17% controls and 11% of FMS
samples, without statistical differences between the groups
(p=0.421).
Furthermore, MCP-1 showed weak or moderate positivity
in all samples, with a significant higher expression in controls
than in FMS patients (p<0.001). More specifically, MCP-1
staining was negative in 4% of controls and in 35% of FMS
patients, and weakly or moderately positive in mastocytes,
appendages, endothelium, and nerves of 96% of controls and
65% of FMS samples.
Similarly, TNFαshowed weak or moderate positivity in
all of our samples. TNFαwas negative in 47% of controls
and in 42% of FMS samples, while it was positive in
mastocytes, appendages, endothelium and nerves of 53%
controls, and in 58% of FMS, with no statistical differences
between groups (p=0.689).
Finally, VEGF showed a low expression in all subjects,
with a higher, marginally significant expression (p=0.051)
in controls than in FMS patients. Particularly, VEGF was
negative in 29% of controls and in 49% of FMS patients,
and weakly positive in endothelium of 27% controls, as
opposed to 14.3% of FMS patients.
No significant differences for any of these above
markers were observed between skin samples from patients
and controls with AAT deficiency phenotypes (MS, MZ,
SZ, and ZZ) and normal (MM) AAT phenotypes.
Calculated inter-rater reliability κvalues for AAT,
tryptase, elastase, MCP-1, TNFα,VEGF,andPAR
2
stains were: 0.84, 0.95, 0.82, 0.65, 0.64, 0.62, and 0.78,
respectively.
Immunohistochemical staining of the above markers are
summarized in Tables 5and 6, and Figs. 1and 2.
AAT genotype Controls Fibromyalgia syndrome p
nMean SD nMean SD
MM 27 146 19.8 39 152.8 13.7 0.102
MS 9 103 6 9 100 5 0.250
MZ 5 70 5 7 67 3 0.231
SS 1 71 0
SZ 3 62 6 6 59 6 0.507
ZZ 4 33 5 2 34 1 0.855
Table 3 Alpha1-antitrypsin
serum concentrations in normal
(MM) and AAT-deficient
genotypes (MS, SS, MZ, SS,
SZ, and ZZ) from 49 controls
and 63 fibromyalgia
syndrome patients
Data given in mg/dL. No
significant differences between
controls and patients found
Table 2 Demographic characteristics (number, age, and gender) of study populations
Samples Subjects number Age mean (standard deviation) [range] Female-to-male ratio (percentage)
General population controls 49 51.1 (7.5) [39–67] 46/3 (93.8/6.2)
Fibromyalgia syndrome patients 63 53.4 (8.7) [34–67] 62/1 (98.4/1.6)
Total 112 52.4 (8.2) [34–67] 108/4 (96.4/3.6)
No significant differences found in ages (p=0.134)
Clin Rheumatol
Discussion
The most striking result of the present study was the finding of
significantly increased numbers of mastocytes in papillary
dermis of all (100%) FMS patients. According to our
immunohistochemistry data, these mastocytes were highly
loaded with tryptase, and contained moderately high levels of
AAT. In our series, FMS patients showed greater than or equal
to five to 14 mastocytes positive for tryptase and AAT per
microscopic high power field (×400) as compared to zero to
one in controls. Most MCs showed a spindle-like (fusiform)
shape, and appeared loosely scattered and uniformly distrib-
uted around dermal blood vessels and appendages of papillary
dermis (Fig 1). These findings are clearly different from
those described in skin mastocytosis, which is characterized
by dense focal infiltrates of tryptase-positive spindle-shaped
MCs (>15 cells/cluster) or scattered MCs exceeding 20 cells
per microscopic high power field [18].
Dermal overexpression of damaged/degranulated masto-
cytes was firstly described in the 1990s by Enestrom et al.
Genotype NMean Std. deviation Minimum Maximum p
Age (years) MM 39 54.5 8.4 34 66 0.210
MS 9 56.3 6.1 45 64
MZ 7 49.0 9.5 39 67
SZ 6 49.3 9.2 38 60
ZZ 2 47.0 16.9 35 59
Total 63 53.4 8.7 34 67
FIQ (scale 0–80) MM 39 65.7 2.8 60 70 0.138
MS 9 66.5 1.0 65 68
MZ 7 64.1 2.7 60 67
SZ 6 66.8 1.6 64 68
ZZ 2 68.5 0.7 68 69
Total 63 65.8 2.6 60 70
HAQ (scale 0–3) MM 39 1.7 0.2 1.4 2.2 0.167
MS 9 1.7 0.2 1.5 2.0
MZ 7 1.7 0.2 1.5 2.0
SZ 6 1.8 0.1 1.7 2.0
ZZ 2 2.0 0.3 1.8 2.2
Total 63 1.7 0.2 1.4 2.2
TePs (scale 0–18) MM 39 16.2 1.3 14 18 0.112
MS 9 17.1 1.0 15 18
MZ 7 16.4 0.5 16 17
SZ 6 16.3 1.2 15 18
ZZ 2 18.0 0.0 18 18
Total 63 16.4 1.2 14 18
Table 4 Demographic and
clinical data of the fibromyalgia
syndrome cohort
No significant differences
between groups found
FIQ Fibromyalgia Impact
Questionnaire, HAQ Health
Assessment Questionnaire,
TePs Tender Points
Table 5 General skin immunohistochemical findings
Antibody Controls (n: 49) Fibromyalgia (n: 63) p
Negative/weak
positive (%)
Moderate/strong
positive (%)
Negative/weak
positive (%)
Moderate/strong
positive (%)
α1-Antitrypsin 98 2 0 100 <0.001
Tryptase 100 0 0 100 <0.001
PAR
2
57.1 42.8 17.5 82.6 <0.001
Elastase 83.7 16.7 88.9 11.1 0.421
MCP-1 4.1 95.9 34.9 65.1 0.000
TNFα46.8 53.2 42.4 57.6 0.689
VEGF 28.6 26.8 49.2 14.3 0.051
PA R
2
proteinase-activated receptors 2, MCP-1 monocyte chemotactic protein-1, TNFαtumor necrosis factor alpha, VEGF vascular endothelial
grow factor
Clin Rheumatol
in a subset of 24 female FMS patients stained with
tolouidine blue. They attributed their findings to a possible
excess of substance P, which can in turn trigger neurogenic
inflammation and MC histamine release [6,7]. However,
nowadays, immunohistochemistry with antibodies against
tryptase has become the most reliable method to stain
mastocytes. Indeed, this technique has demonstrated to be a
highly sensitive, specific, and reproducible procedure for the
recognition of even hypogranulated, non-metachromatic, very
atypical or immature mastocytes, which can be used for the
detection of even minute infiltrates that can be easily
overlooked when using other methods [19,20].
Our study also showed moderate or strong positive
staining for AAT in papillary dermis mastocytes with less
frequency and intensity than tryptase. This finding has been
previously described and attributed to the high active serine
proteinase expression by tissue MCs. AAT is thought to be
normally produced in the tissue to prevent the proteolytic
action of these proteinases [21].
Other interesting findings of the current study were: (1) a
significant overexpression of PAR
2
in mastocytes and in
other papillary dermis components of FMS patients, and (2)
a decreased expression of MCP-1 and VEGF in the skin of
FMS patients. PAR
2
expression in skin of FMS patients has
never been studied before, although its possible participa-
tion in a pain pathway mediated by mastocytes/PAR
2
in
FMS has been recently hypothesized [22]. Neither the
expression of MCP-1 and VEGF in skin of FMS has
been previously studied, but these biomarkers have been
found to be decreased in blood samples from FMS
patients [3]. The remaining antibodies used for immuno-
histochemistry in our study (i.e., for elastase, and TNFα)
Fig. 1 Representative tryptase
staining (dark brown)in
papillary dermis from paraffin-
embedded samples of fibromy-
algia patients. aMast cell
tryptase
+
in capillary vessels
and appendages (×250). band c
mast cell tryptase
+
in capillary
(×400). dMast cell tryptase
+
in
appendages (×400). The arrows
denote representative sites of
staining
Table 6 Numbers and location of MC-tryptase
+
, MC-α1-antitrypsin
+
, and MC-PAR
2+
in controls and FMS patients
Antibody Controls (n: 49) Fibromyalgia (n: 63) p
MC numbers and location in papillary dermis MC number and location in papillary dermis
≤1 PV or NPV (%) ≥5–14 PV (%) ≥5–14 NPV (%) ≤1 PV or NPV (%) ≥5–14 PV (%) ≥5–14 NPV (%)
Tryptase 100 0 0 0 39.7 60.3 <0.001
α1-Antitrypsin 98 2 0 0 47.6 52.4 <0.001
PAR
2
57.1 30.6 12.2 17.5 65.1 17.5 <0.001
MC mast cells, PV perivascular mast cells, around small vessels, NPV non-perivascular mast cells but other papillary dermis components, mostly
epidermal appendages (hair follicles, sebaceous and sweat glands) and, occasionally, extracellular matrix, small muscles, fat, and nerve tissues,
PA R
2
proteinase-activated receptors 2
Clin Rheumatol
did not prove to be of use for ‘biomarker’detectioninthe
setting of FMS.
A complementary observation of the present study was
that all our patients were collected using the 1990 ACR
criteria, after reasonably excluding other diseases mimicking
fibromyalgia, and that they invariably showed increased
numbers of dermis MCs. Therefore, the usually utilized
clinical approach for the selection of FMS patients, based on
the 1990 ACR criteria, appears to be highly sensitive, and
should be strongly recommended in order to make an initial
clinical diagnosis of the disease.
Another observation of the present study worthy of
mention was that we did not find any gross immunohisto-
chemical differences between AAT deficiency and non-
AAT deficiency FMS samples; all biopsies showed similar
and indistinguishable histological findings at the gross
histological level. Anecdotally, the only FMS male of our
series also showed immunohistochemical changes similar
to those seen in FMS females. Also, it should be mentioned
that the skin samples from the three FMS/AAT-deficient
females who received long-term AAT-replacement therapy
and who had good clinical control of their FMS symptoms
[15,17], showed a moderate increase of the number of
dermal MCs (approximately eight to 14 cells/field). This
result suggests that AAT is a very good therapeutic agent to
control FMS symptoms, regardless of its ability to reverse
the abnormally increased number of skin MCs.
The dermis is a complex structure composed of two layers:
the more superficial is the papillary dermis and the deeper, the
reticular dermis. Dermis contains an extracellular matrix
primarily composed of glycosaminoglycans (especially hya-
luronan), proteoglycans, and glycoproteins. Other compo-
nents of the papillary dermis are collagen and elastin fibers,
cells (e.g., fibroblasts, histiocytes, T-cells, and MCs), blood
and lymphatic vessels, small muscles, fat, nerve tissues
(including mechanoreceptors, nociceptors, and thermorecep-
tors), and epidermal appendages (i.e., sebaceous, sweat,
apocrine, and mammary glands, and hair follicles). MCs are
normal connective tissue components, widely distributed at
body sites close to the surface (i.e., skin and mucosal surfaces
of the lung, digestive tract, conjunctiva, mouth, and nose),
where they act as a first-line defense of innate immunity
against invasion by external pathogens. Mast cells are derived
from a multipotent stem cell in bone marrow, circulate in the
blood as immature precursors, and migrate into tissues in
which they undergo terminal differentiation under the
influence of local growth factors, in particular the stromal-
cell-derived cytokine stem cell factor (SCF), acting as a ligand
of the mastocyte cell membrane tyrosine kinase receptor
(c-KIT or CD117). SCF is mainly produced by fibroblasts, but
also by histiocytes, endothelial cells, and activated mature
MCs. In neoplastic states, the autonomous MC proliferation
can be induced by activating mutations of c-KIT, mainly the
gain-of-function mutation in codon 816 of c-KIT frequently
present in MCs of patients with systemic mastocytosis [23].
Mature MCs develop hundreds of cytoplasmic vesicles,
containing histamine (2–5 pg/cell), serotonin, heparin,
cytokines (i.e., tumor necrosis factor, and interleukins 4, 5,
Fig. 2 Representative staining
(brown) in papillary dermis from
paraffin-embedded samples of
fibromyalgia patients. aand b
Mast cell alpha1-antitrypsin
+
around vessels (×400). cand d
Mast cell PAR
2+
in capillary
vessels (×400). The arrows
denote representative sites
of staining
Clin Rheumatol
6, and 8), eosinophil chemotactic factor, basic fibroblast
growth factor, SCF, endothelial permeability factor, VEGF,
serine proteinases (i.e., tryptase and chymase), and the zinc-
dependent metalloproteinase MC carboxypeptidase A (MC-
CPA). The amount of proteinase stored in MC vesicles is not
only very high (16 mg of tryptase and chymase per 10
6
human
skin MCs) but also highly specific for these cells, such that
MCs can be readily distinguished immunohistochemically
from other cells in tissues. In addition, the specific
expression pattern for MC proteinases is useful to determine
the subclass of MCs in humans, with MCs of the “MC T”
class (mainly present in mucosal surfaces) expressing only
tryptase, and the “MC TC”subclass (very abundant in skin
and intestinal submucosa) expressing tryptase, chymase, and
MC-CPA [24]. Mastocytes can be stimulated by physical or
chemical injuries, immunoglobulin E (IgE), pathogen-
derived products (e.g., LPS, lipopeptides, flagellin, or viral
RNA and DNA), C3a and C5a complement, endothelin 1
(ET-1), and substance P to degranulate and release their
mediators into the interstitium. This mastocyte activation
promotes the synthesis and release of newly formed lipid-
membrane-derived eicosanoids, like prostaglandin D2
(PGD2), leukotriene B4 (LTB4), leukotriene C4 (LTC4),
leukotriene D4 (LTD4), and platelet-activating factor (PAF).
In this process, several of the released substances can act in
different components of the dermis, including MC receptors
(e.g., histamine through H1 receptors, or tryptase through
PAR
2
), inducing further activation/degranulation of adjacent
MCs and amplifying initial activation effects. At an early
stage, proteinases can degrade normal dermis components.
Subsequently, proteinase inhibitors escaping from plasma to
reach the neighboring tissues down-regulate proteinase
activity. “In vitro”AAT strongly inhibits induced histamine
and tryptase release from human MCs in a concentration-
dependent manner [25,26], and facilitates the repair of
damaged connective tissue by stimulating fibroblast prolif-
eration and extracellular matrix production via the classical
mitogen-activated signaling pathways [27]. However, little is
known about the precise in vivo mechanisms that modulate
the activity of inflammatory mediators, which probably are
under stringent control to prevent host tissue damage. In fact,
alpha2-macroglobulin and the serpins AAT, the secretory
leukocyte proteinase inhibitor (SLPI), and the squamous cell
carcinoma antigen 2 are efficient inhibitors of chymases, and
after exocytosis, the tryptase tetramer dissociates into
monomers that can be targeted by endogenous proteinase
inhibitors that otherwise fail to inhibit tryptase when it is in
its tetrameric form [24].
Since skin is the most extensive organ of the human body
with around 2 m
2
and 7,000 to 10,000 MCs in 1 mm
3
,in
FMS (an apparently dermis mastocyte-associated disease)
the number of mastocytes is increased up to 14 times and,
therefore, mastocyte-released chemicals could cause local
manifestations (e.g., skin tenderness and pain) and systemic
symptoms through the blood. Moreover, this increased
amount of activated MCs dermis mediators can be the source
of increased peripheral stimuli reaching CNS structures which
cause central hypersensitivity. In skin mastocytosis, massive
mastocyte-released mediators can cause local skin manifes-
tations, while they can also disseminate through the circula-
tion resulting in the manifestation of several systemic
symptoms, mainly fatigue, headache, flushing, abdominal
discomfort, nausea, hypotension, and tachycardia. In systemic
mastocytosis, serum levels of tryptase greater than 20 ng/mL
can be usually found, serving as a clinical indicator of disease
activity. Other mastocyte-related molecules including hista-
mine and its metabolites, as well as soluble KIT or soluble
CD25, can also serve as surrogate markers, although their
contribution when combined with tryptase remains to be
determined. None of these mastocyte-related substances have
been tested in FMS patients so far.
In summary, the main results of the present study can lead
to the following conclusions: (1) FMS seems to be a papillary
dermis disease associated with increased numbers of masto-
cytes; (2) the 1990 ACR criteria should be strongly supported
for a clinical diagnosis of suspected cases of FMS; (3)
although clinical and epidemiological data suggest that FMS
can be related to Z allele of AAT deficiency, no significant
differences between AAT deficiency and non-AAT deficiency
subjects have been found in our skin biopsies; (4) decreased
expression of MCP-1 and VEGF in blood and skin of FMS
patients can indicate a dysregulation of these markers in FMS
patients; (5) mastocyte degranulation-related symptoms (such
as fatigue, headache, flushing, abdominal discomfort, hypo-
tension, and tachycardia) are frequently present in FMS
patients, supporting the association of the disease with
mastocytes degranulation and release of active mediators.
Our findings are thus highly relevant (1) for the future
management of FMS patients, in terms of their diagnosis and
therapy, and (2) for pointing to new areas of researchrelated to
the pathogenesis of FMS.
Acknowledgments We would like to thank every patient and volunteer
who participated in this study, providing us with their clinical data and
biopsy samples. We also acknowledge the technical work for the
preparation of the paraffin-embedded tissues samples by Dr. Francisco
Domínguez and laboratory technicians at the Valle del Nalón Hospital,
Asturias, Spain; and the immunohistochemical work performed by Ms.
Mercedes Acha (Cabueñes Hospital, Gijón, Spain). Furthermore, we are
indebted toDr. Eleftherios P. Diamandis (Mount Sinai Hospital, Toronto)
for sharing vital reagents and ideas for the completion of this work.
Finally, we are grateful to Ms. Jimena Blanco Fueyo (UNESCO; MA/BA
Université de Genève) for the English editing of this manuscript. This
study has been endorsed by the Spanish National Health Institute Carlos
III and the Biohealth Research Office (OIB) of the Principado de Asturias,
Spain (IB and VC). KO is a recipient of an Alberta Heritage Foundation
for Medical Research (AHFMR) Postdoctoral Fellowship for whom
operating funds were provided by a grant to MDH by the Canadian
Institutes of Health Research.
Clin Rheumatol
Disclosures None
References
1. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C,
Goldenberg DL, Tugwell P, Campbell SM, Abeles M, Clark P,
Fam AG, Farber SJ, Fiechtner JJ, Franklin CM, Gatter RA,
Hamaty D, Lessard J, Lichtbroun AS, Masi AT, McCain GA,
Reynolds WJ, Romano TJ, Russell IJ, Sheon RP (1990) The
American College of Rheumatology 1990 criteria for the
classification of fibromyalgia. Arthritis Rheum 33:160–172
2. Staud R (2002) Evidence of involvement of central neural
mechanisms in generating fibromyalgia pain. Curr Rheumatol
Rep 4:299–305
3. Blanco I, Janciauskiene S, Nita I, Fernández-Bustillo E, Cárcaba
V, Gallo C, Alvarez-Rico M, de Serres F, Béridze N (2010) Low
plasma levels of monocyte chemoattractant protein-1 (MCP-1),
tumor necrosis factor-alpha (TNFalpha), and vascular endothelial
growth factor (VEGF) in patients with alpha1-antitrypsin
deficiency-related fibromyalgia. Clin Rheumatol 29(2):189–197
4. Blanco I, Arbesú D, Al Kassam D, de Serres F, Fernández-
Bustillo E, Rodríguez Menéndez C (2006) Alpha1-antitrypsin
polymorphism in fibromyalgia patients from the Asturias Province
in Northern Spain: a significantly higher prevalence of the PI*Z
deficiency allele in patients than in the general population. J
Musculoskelet Pain 14(3):5–12
5. Eneström S, Bengtson A, Lindström F, Johan K (1990) Attachment
of IgG to dermal extracellular matrix in patients with fibromyalgia.
Clin Exp Rheumatol 8(2):127–135
6. Enestrom S, Bengtsson A, Frodin T (1997) Dermal IgG deposits
and increase of mast cells in patients with fibromyalgia—relevant
findings or epiphenomena? Scand J Rheumatol 26(4):308–313
7. Salemi S, Rethage J, Wollina U, Michel BA, Gay RE, Gay S, Sprott
H (2003) Detection of Interleukin 1 beta (IL-beta), IL-6, and tumor
necrosis factor-alpha in skin of patients with fibromyalgia. J
Rheumatol 30(1):146–150
8. Sprott H, Muller A, Heine H (1997) Collagen crosslinks in
fibromyalgia. Arthritis Rheum 40(8):1450–1454
9. Ribel-Madsen S, Gronemann ST, Bartels EM, Danneskiold-
Samsoe B, Bliddal H (2005) Collagen structure in skin from
fibromyalgia patients. Int J Tissue React 27(3):75–82
10. Kim SH, Jang TJ, Moon IS (2006) Increased expression of
N-methyl-D-aspartate receptor subunit 2D in the skin of patients
with fibromyalgia. J Rheumatol 33(4):785–788
11. Kim SH (2007) Skin biopsy findings: implications for the
pathophysiology of fibromyalgia. Med Hypotheses 69(1):141–144
12. Kim SH, Kim DH, Oh DH, Clauw DJ (2008) Characteristic
electron microscopic findings in the skin of patients with
fibromyalgia—preliminary study. Clin Rheumatol 27(3):407–411
13. Janciauskiene S, Stevens T, Blanco I (2007) New insights into the
biology of α1-antitrypsin and its role in chronic obstructive lung
disease. Respir Res 3(2):147–158
14. Stoller JK, Snider GL, Brantly ML (2003) American Thoracic
Society/European Respiratory Society Statement: standards for
the diagnosis and management of individuals with alpha-1
antitrypsin deficiency. Am J Respir Crit Care Med 168(7):818–
855
15. Blanco I, Canto H, de Serres FJ, Fernandez-Bustillo E, Rodriguez
MC (2004) Alpha1-antitrypsin replacement therapy controls fibro-
myalgia symptoms in 2 patients with PI ZZ alpha1-antitrypsin
deficiency. J Rheumatol 31(10):2082–2085
16. Blanco I, de Serres FJ, F-Bustillo E, Al Kassam D, Arbesú D,
Rodríguez C, Torre JC (2005) Alpha 1-Antitrypsin and fibromy-
algia: new data in favour of the inflammatory hypothesis of
fibromyalgia. Med Hypotheses (Oxford) 64:759–769
17. Blanco, A Astudillo, F Domínguez, S Janciauskiene, V Cárcaba,
C Gallo, H Canto, F de Serres, E F-Bustillo (2010) Intravenous
infusions of purified alpha 1-antitrypsin effectively controls
symptoms and reverts muscle biopsy changes in an MZ-alpha-1
antitrypsin deficiency and fibromyalgia syndrome Patient. J
Musculoskel Pain (in press)
18. Horny HP, Valent P (2001) Diagnosis of mastocytosis: general
histopathological aspects, morphological criteria, and immunohis-
tochemical findings. Leuk Res 25(7):543–551
19. Li WV, Kapadia SB, Sonmez-Alpan E, Swerdlow SH (1996)
Immunohistochemical characterization of mast cell disease in
paraffin sections using tryptase, CD68, myeloperoxidase, lysozyme,
and CD20 antibodies. Mod Pathol 9(10):982–988
20. Valent P, Horny HP, Escribano L, Longley BJ, Li CY, Schwartz
LB, Marone G, Nuñez R, Akin C, Sotlar K, Sperr WR, Wolff K,
Brunning RD, Parwaresch RM, Austen KF, Lennert K, Metcalfe
DD, Vardiman JW, Bennett JM (2001) Diagnostic criteria and
classification of mastocytosis: a consensus proposal. Leuk Res 25
(7):603–625
21. Benítez-Bribiesca L, Freyre R, De la Vega G (1973) Alpha-1-
antitrypsin in human mast cells immunofluorescent localization.
Life Sci 13(6):631–638
22. Masuko K, Nakamura H (2007) Functional somatic syndrome:
how it could be relevant to rheumatologists. Mod Rheumatol 17
(3):179–184
23. Bunimovich O, Grassi M, Baer MR (2009) Systemic mastocytosis:
classification, pathogenesis, diagnosis, and treatment. Cutis 83(1):29–
36
24. Pejler G, Åbrink M, Ringvall M, Wernersson S (2007) Mast cell
proteases. Adv Immunol 95:167–255
25. He SH, Xie H, Zhang XJ, Wang XJ (2004) Inhibition of histamine
release from human mast cells by natural chymase inhibitors. Acta
Pharmacol Sin 25(6):822–826
26. He SH, Xie H (2004) Inhibition of tryptase release from human
colon mast cells by protease inhibitors. World J Gastroenterol 10
(3):332–336
27. Dabbagh K, Laurent GJ, Shock A, Leoni P, Papakrivopoulou
J, Chambers RC (2001) Alpha-1-antitrypsin stimulates fibro-
blast proliferation and procollagen production and activates
classical MAP kinase signalling pathways. J Cell Physiol 186
(1):73–81
Clin Rheumatol
